The present invention relates to a method and arrangement for detecting and preferably tracking a non-authorised client stations attempt to access a host in a communications network. The communication network comprises connection devices at least each having a unique identity.
The invention also relates to a telecommunications network having a security arrangement.
During recent years, the art of digitalised communications using for example computers, mobile phones etc., has changed dramatically. The changes, besides of being an aid for users and allowing faster and better communication, have given some people opportunity to use the benefits and possibilities of essentially advanced communication means to carry out, more or less, criminal acts, such as fraudulence, e.g. by accessing the company or government computer networks and retrieving, changing or deleting information, or using telephones, switchboards etc. to obtain felonious privileges.
Internet, the global network for computers and computer networks have also contributed to the global communications, by allowing transference of images, voices and other data in a simple and inexpensive way. The result is that companies, government offices, universities and so on, have connected their networks through Internet to supply the internal and external users with relevant information and also to communicate with each other.
As the Internet is a public network, i.e. everybody having a computer and a communication device, such as a modem, may through an Internet Provider Server (IPS) access the Internet and communicate with others or just retrieve information. Internet is, in a superior manner, the fastest and most effective way to distribute a large amount of information for a large number of people.
The "core" of the Internet includes a very large number of computers, standalone or connected in networks, which can exchange information substantially directly using some predetermined protocols, especially Transmission Control Protocol/Internet Protocol (TCP/IP).
Each computer or connecting device in the core is separated from each other by means of an IP address. The IP address consists of a network number. In some cases the IP address is permanently assigned to a device (computer) in other cases the IP address is assigned to a computer temporarily. The IP address provides each connected computer/device a unique identity in the network.
The data transmission may be carried out via, for example fiber-optic lines, satellite links and telephone lines.
At present, obtaining full identification of a client or a user workstation connected or trying to connect to a host system is not possible. Also, It is not possible, in a simple and fast way, to identify the fraud workstation and thereby the user, even though a partial identification is possible.
When accessing a network, usually a login procedure is executed for authentication of the user. The authentication works, by the client first declaring the user name to be used to access the network. The service providing server then responds with a set of authentication methods, which are acceptable. The client then sends an authentication request, and this dialogue continues until an access has been granted or denied. The authentication methods can vary from system to system. Some methods are:
none checks if no authentication is ok, password a conventional password authentication, which requires a password for access, secureid secureID authentication is a timing-based hardware token identification, where the user enters a code displayed on the token as authentication.
Also, on-time passwords and similar methods are available. As other methods public key can be mentioned, in which the possession of a privet key is the authentication.
Great efforts are made to develop methods and algorithms to secure the authentication procedures, but no system is more secure than the user of the system, as passwords and keys may come in possession of non authorised persons. In the Internet case, also the IP addresses can be forged which allows accessing the network without problem.
By forged IP address, it is also possible to attack the Internet service providers, for example by flooding or the like. Flooding is a method where an unreachable source (IP) address is used against a target host computer, which attempts to reserve resources waiting for a response. The attacker repeatedly changes the bogus source address on each new access packet sent, thus exhausting additional host resources. Then, if the attacker uses some valid address as the source address the attacked system responds by sending a large number of reply packages, which at end, results in a degraded performance and even system crash.
U.S. Pat. No. 5,619,657 teaches a method for providing security facility for a network of management servers utilising a database of trust relations to verify mutual relations between management servers. The method relates specially to creating accounts on a system over a network. The management operation (MO) also contains the identity of the user lunching the operation. Through an interface the MO is transferred to a dispatcher of the management server (MS). The MS, in addition to administrating, requests for management services provided by a local system is also responsible for routing MOs on secure paths to other local systems in the network and managing the security of the local system. The MS determines a proper link by means of a database, which maintains trusted relations between the management servers. The trusted relation lists are generated independent from the execution of a communication protocol by an autonomous network utility. Each MS contains a list. The lists are divided into two categories, trusted receivers and trusted senders. Based on the trusted list, forwarding the operations are executed. In summary, the database provides a means for routing MOs from one MS to another MS along a secure path determined by the trust relations of the MSs at each link in the route in the network performing the MO.